Jeopardy, Consciousness, and Multiple Discrimination: Intersecting Inequalities in Contemporary Western Europe

Catherine E. Harnois1

Theories of intersectionality argue that individuals with multiple minority statuses often face mistreatment that stems from multiple, interlocking systems of inequality. King (1988) refers to this phenomenon as “multiple jeopardy,” and argues that those who experience multiple jeopardy often develop a “multiple consciousness”—an awareness of multiple systems of inequality working with and through one another. This study analyzes recent survey data to assess perceived multiple jeopardy and its relationship to multiple consciousness in the context of contemporary Western Europe. Findings provide support for intersectionality, as individuals who hold multiple minority statuses are more likely than others to perceive having personally experienced multiple forms of discrimination, and are more likely to view multiple discrimination (discrimination based on multiple social statuses) as a widespread social phenomenon. Controlling for other factors, personal experiences with multiple forms of discrimination (“multiple jeopardy”) are associated with greater multiple consciousness. Personal experiences with discrimination based on a single dimension of inequality (“single jeopardy”) also facilitate multiple consciousness, however, though not to the same degree. The conclusion highlights the importance of intersectionality for future research and policy concerning discrimination.

KEY WORDS: discrimination; ethnicity; Europe; gender; intersectionality; survey research.

INTRODUCTION

Over the past two decades, the concept of intersectionality has become increasingly central to sociological scholarship, as well as to social science research more generally. The particulars of how scholars understand and employ intersectionality vary, but, in general, intersectionality draws attention to the dynamic and contextual relationships among multiple systems of inequality, as they exist within and across multiple levels of society (Baca Zinn and Thornton Dill 1996; Berger and Guidroz 2009; Choo and Ferree 2010; Hill Collins 2000; McCall 2005; Weber 2010; Winker and Degele 2011; Yuval-Davis 2006). While numerous studies have demonstrated that intersectionality provides a valuable framework for understanding how intersecting inequalities structure disadvantage and privilege, far fewer have investigated the social psychological claims central to intersectionality. Intersectionality theory argues that individuals who have multiple minority statuses often experience mistreatment that stems from multiple, interconnected systems of inequality (Baca Zinn and Thornton Dill 1996; Crenshaw 1989, 1991; Hill Collins 2000; King 1988; Weber 2010). Intersectionality This is an open access article under the terms of the Creative Commons Attribution-NonCommercial- NoDerivs License, which permits use and distribution in any medium, provided the original work is properly cited, the use is non-commercial and no modiﬁcations or adaptations are made. 1 Department of Sociology and Gender, Women’s and Sexuality Studies, P.O. Box 7808, Wake Forest University, Winston-Salem, North Carolina 27109; e-mail: [email protected].

971 972 Harnois further posits that those who experience mistreatment based on multiple social statuses often recognize it as such, framing their experiences in the context of multiple social statuses and multiple systems of inequality (Crenshaw 1989, 1991; Essed 1991; Guenther, Pendaz, and Makene 2011; Harnois 2014; Hill Collins 2000; King 1988; Risman and Banerjee 2013). Finally, intersectionality theorists have argued that those who experience mistreatment based on multiple social statuses often develop a broader understanding of intersecting social inequalities (Berger 2004; Hill Collins 1990; King 1988; Smith, Smith, and Frazier 1981). This study addresses the social psychological claims of intersectionality directly, asking two main questions. First, to what extent do individuals with multiple minority statuses understand their experiences with mistreatment in terms of multiple inequalities? And second, to what extent does membership in multiple minority groups, along with experiences of multiple forms of discrimination, facilitate a broader awareness of intersecting inequalities? To answer these questions, I analyze data from the Eurobarometer Survey—a multicountry survey conducted regularly on behalf of the European Commission. While intersectionality was developed primarily by women of color in the United States, the theoretical implications of intersectionality reach far beyond this particular context. Unlike any large-scale survey conducted in the United States, the 2008 Eurobarometer includes questions about multiple minority identities, multiple forms of discrimination, and most sig- niﬁcantly, a question that asks directly about respondents’ perceptions of “multiple discrimination”—discrimination based on multiple social statuses. It represents one of the only existing large-scale surveys—if not the only large-scale survey—with which it is possible to assess the prevalence of perceived multiple jeopardy in any context, and its relationship to multiple consciousness.

BACKGROUND Multiple Jeopardy and Multiple Consciousness

More than 25 years ago, Deborah King’s (1988) article, “Multiple Jeopardy, Multiple Consciousness: The Context of a Black Feminist Ideology,” challenged scholars and activists to rethink their approach to understanding and addressing inequalities of race, class, and gender. King argued that scholars and activists alike too often rely on a “monist” approach, which obscures the experiences of individuals and groups who hold multiple minority statuses. Drawing on Albert et al. (1986), she described a monist approach as one that positioned a single inequality (e.g., race, class, or gender) as foundational to others, and that conceptualized systems of inequality as occurring independently (King 1988:51). King argued that attempts to isolate the separate contributions of racism, classism, and sexism to African American women’s lives fundamentally misunderstood how systems of inequality operate and are experienced. She advocated for an alternative approach, in which systems of inequality were viewed as dynamically interacting with one another. King’s theory draws from the scholarship, politics, and everyday experiences of African American women, and documents that throughout history African Intersecting Inequalities in Western Europe 973

American women’s lives have been structured by multiple, intersecting systems of inequality. She introduces the concept of “multiple jeopardy” to describe the interpersonal mistreatment and structural disadvantage that result from African Ameri- can women’s multiple minority statuses. The term multiple jeopardy builds upon the “additive” models of inequality that came before (e.g., Beal 1970), and “refers not only to several, simultaneous oppressions, but to the multiplicative relationships among them as well” (King 1988:47). For King, and for much of the intersectional scholarship that has followed, “the intersection of racism and sexism factors into Black women’s lives in ways that cannot be captured wholly by looking at the race or gender dimensions of those experiences separately” (Crenshaw 1991:1244; see also Baca Zinn and Thornton Dill 1996; Bowleg 2008; Essed 1991; Hill Collins 2000). In addition to highlighting the experience of multiple jeopardy, King argues that African American women’s experiences with multiple jeopardy facilitate the development of “multiple consciousness”—an awareness of multiple systems of inequality, and an awareness of how these systems “work with and through each other” (Baca Zinn and Thornton Dill 1996). She posits that those who occupy multiple marginal statuses, and whose lives are characterized by multiple jeopardy, have a unique standpoint, which facilitates an understanding of how inequalities work together to structure power and privilege. Drawing from the writings of Anna Julia Cooper, Mary Church Terrell, Angela Davis, Audre Lorde, and others, King (1988:51) argued that historically, black women in the United States have been conscious of multiple systems of domination and have understood these interacting systems as “a primary inﬂuence on the black woman’s deﬁnition of her womanhood, and her relationships to people around her” (see also Hill Collins 2000; Smith et al. 1981; Wilcox 1997). While King’s theory of multiple jeopardy and multiple consciousness central- ized the experiences of African American women and focused on inequalities of race, gender, and class, more recent scholarship has used these concepts to examine a broader range of inequalities as they relate to a wide range of marginalized groups around the world. For example, Cooke (2001) draws on the concepts of multiple jeopardy and multiple consciousness to describe the “oppositional stance” of post- colonial Arab Muslim women, Ward (2004) uses these concepts to examine the experiences of lesbian women at a Latino health clinic in Los Angeles, and Madibbo (2007) applies these ideas to blacks in the Francophone communities of Ontario, Canada. In addition to the statuses of race, class, and gender, scholars have drawn on King’s theory to understand inequalities related to immigration status, (dis)ability, sexual orientation, age, and religion (e.g., Harnois 2014; Sargeant 2011; Ward 2004; Weber 2010).

The Correlates of Perceived Discrimination

No existing large-scale statistical study has examined explicitly the relationship between multiple jeopardy and multiple consciousness, in any context. Existing social science research on discrimination is useful, however, for considering the 974 Harnois potential correlates of multiple jeopardy and multiple consciousness. Previous research in the U.S. context has found that racial-ethnic minorities (and blacks in particular) are more likely to perceive having experienced racial-ethnic discrimination than are non-Hispanic whites (Kessler, Mickelson, and Williams 1999; Welch et al. 2001). Younger and older adults are more likely than middle-aged adults to perceive high levels of age-based discrimination (Garstka, Hummert, and Bran- scombe 2005; Gee, Pavalko, and Long 2007). Women are more likely than men to report experiencing gender-based discrimination and sexism (Kobrynowicz and Branscombe 1997; Swim et al. 2001, but racial-ethnic minority men tend to report higher levels of racial-ethnic discrimination than do racial-ethnic minority women.2 Some studies suggest higher educational attainment increases the likelihood of perceiving discrimination (Gee et al. 2007; Kessler et al.1999; Sigelman and Welch 1991). Workforce participation and occupational sector may also shape the likelihood of perceiving discrimination (Roscigno 2007; Roscigno et al. 2007). An important limitation to the research described above is that existing studies tend to focus on either discrimination “in general” (e.g., Kessler et al. 1999) or on one specific type of discrimination (e.g., Gee et al. 2007; Kobrynowicz and Bran- scombe 1997). Studies focusing on discrimination “in general” make no distinction between single and multiple discrimination, and are thus limited in their ability to assess potential differences between the two. Studies focusing on a single type of discrimination are similarly limited in their ability to assess multiple discrimination. Often relying upon a “monist” framework, these studies imply that various types of discrimination generally occur separately, and that those who experience racial, age-based, and/or gender-based discrimination interpret this mistreatment as stemming from a single system of inequality (Harnois 2014; Harnois and Ifatunji 2011). In contrast to these studies, recent research by Best and colleagues (2011) analyzes a sample of judicial opinions in equal employment opportunity cases in the U.S. courts from 1965 through 1999, and finds that in 18% of cases plaintiffs made claims based on multiple statuses. In these cases, the most frequent combination was a claim of gender and race discrimination (6% of the sample), followed by race and national origin, sex and age, and age and national origin (each approximately 3% of the sample). Significantly, they find that those who made “intersectional claims” were only half as likely to win their cases as those who made claims based on a single social status. Best and colleagues’ research focuses on official claims of discrimination in the context of the U.S. workplace. With an explicit focus on multiple discrimination, and an examination of discrimination as it occurs within and outside of the workplace, the current study expands the scope of research on how discrimination is perceived by those with multiple minority statuses.

Discrimination and Intersectionality in Contemporary Western Europe

As described above, King’s theory of intersectionality emphasizes that the relationships created by the intersection of multiple systems of inequality are dependent

2 Some of this gender difference may be due to gender bias in measurement instruments and modeling techniques (Harnois and Ifatunji 2011). Intersecting Inequalities in Western Europe 975 upon the “socio-historical context and the social phenomenon under consideration” (King 1988:49; see also Hill Collins 2000). King does not argue that every combination of minority statuses will result in multiple jeopardy and the development of a multiple consciousness. Twenty-five years of subsequent intersectional scholarship, however, suggests that the concepts and relationships that King describes may reflect more general processes. To be clear, the broader claim of intersectionality is not that multiple systems of inequality all matter equally for all groups in all situa- tions, but rather, that it is always important to consider the possibility of multiple, intersecting systems of inequality. In any particular social and historical context, multiple systems of inequality may come together to produce multidimensional experiences and a consciousness of intersecting inequalities. Because context is central to both intersectionality and analyzing discrimination, it is important to emphasize that the context in which King’s theory emerged is very different from the context of contemporary Western Europe. In her recent work on the “transatlantic travels” of intersectionality theory, Knapp (2005) highlights the risks of importing theories and concepts from one context and blindly deploying them elsewhere. The “conceptual triad” of race-class-gender—which often works as the sine qua non of intersectionality in the U.S. context—may, for example, actually stifle the potential of intersectionality in German-speaking Eur- ope, where the concept of Klasse (class) retains a strong connection to Marxism, and Rasse (race) “is a category that cannot be used in an affirmative way” (Knapp 2005:257). Rather than rejecting the theory, however, Knapp calls for an intersectionality that is more contextually flexible and attentive to national, regional, and global structures and histories (e.g., European integration, globalization).3 While a complete comparison of the U.S. and Western European sociodemographic, legal, and cultural contexts is beyond the scope of this article, below I highlight some of the most important differences. First among these is that the Western European countries investigated here differ significantly from the United States— and from each other—in terms of their racial-ethnic and religious composition. In the United States, for example, Hispanics and African Americans constitute the largest racial-ethnic minority groups: approximately 16% and 13% of the U.S. population, respectively (U.S. Census Bureau 2011). Western European countries are generally characterized by less racial-ethnic diversity, and the racial-ethnic groups present in these countries reflect particulars of geography, politics, and historical legacies of colonialism.4,5 Turks constitute the largest racial-ethnic minority group in Germany, for example, where they represent 2.4% of the country’s overall population, and 24.4% of the foreign population living in Germany (CIA 2009; Federal Statistical Office of Germany 2011). The Central Intelligence Agency (CIA) estimates that in the United Kingdom, 2% of the population is black, 1.8% is ethni- cally Indian, and 1.3% Pakistani. While the United States is generally characterized

3 See also Ferree (2009) and Choo (2012). 4 An exception is Belgium, where the Walloon ethnic group comprises 31% of the population (CIA 2009). 5 Direct comparison of the racial-ethnic diversity in European countries and the United States is compli- cated by the fact that racial-ethnic categories recognized in these countries differ dramatically, and that countries differ in terms of what demographic information they collect. For this reason I rely on estimates provided by the U.S. CIA Factbook. 976 Harnois as having a higher level of racial-ethnic diversity, Western European countries have relatively high levels of religious diversity, particularly when non–Judeo-Christian religions are considered. While only 0.6% of the U.S. population identifies as Mus- lim, for example, recent estimates suggest that 5.8% of the German population identifies as Muslim and in France this figure is estimated to be between 5% and 10% (CIA 2009; Hackett 2015). In addition to sociodemographic and religious differences, Western European countries differ widely from the United States, and from each other, in their antidiscrimination and antiharassment laws. In the United States, municipalities and states can enact laws prohibiting particular forms of discrimination, but the main institutions responsible for monitoring and addressing discrimination (e.g., the Equal Employment Opportunity Commission, the Department of Housing and Urban Development) operate at the federal level. U.S. federal law currently prohibits employment discrimination on the basis of age, national origin, pregnancy, race/ color, religion, disability, or sex. The Civil Rights Act of 1968 prohibits housing discrimination based on race, color, national origin, religion, sex, familial status, and disability (U.S. HUD 2015). U.S. law does not currently protect against discrimination based on sexual orientation, nor does it protect victims of workplace age discrimination who are under the age of 40 (U.S. EEOC 2010). Until 1997, European antidiscrimination policy also occurred primarily at the national level for each country, but this changed significantly with the 1997 Amsterdam Treaty of the European Union. While some protections against discrimination on the basis of nation and gender had been in place previously, the Amsterdam Treaty, along with the European Union (EU) Racial Equality and Employment Framework Directives of 2000 that followed, added new provisions to combat discrimination based on race, ethnicity, religion, disability, age, and sexual orientation (Amiraux and Guir- audon 2010). Laws concerning public expression also differ in the European and U.S. context, and have important implications for analyzing perceived discrimination. As Bleich (2011) explains, while the U.S. Constitution protects freedom of speech, including the freedom to express racist, anti-Semitic, Islamophobic, and misogynist sentiments, European countries have increasingly legislated against racist speech and the incitement to racial, ethnic, and religious hatred, and other forms of racist expression. In the United Kingdom, for example, the Public Order Act of 1986, the Racial and Religious Hatred Act of 2006, and the 2008 Criminal Justice and Immi- gration Act each work to criminalize the incitement of hatred. Belgian law “penal- izes public announcements of intention to discriminate, hate, or perpetrate violence against an individual or group because of race, color, origin, descent or national- ity,” and Sweden, Denmark, Italy, and France similarly restrict the public expression of racism, ethnocentrism, and religious superiority (Bleich 2011:21).6 These laws make public hate speech against protected groups criminally punishable and, compared to the United States, may reduce the likelihood that minorities may experience explicit discrimination or harassment.

6 Denmark includes mistreatment based on “sexual inclination” as well (Bleich 2011:21). Intersecting Inequalities in Western Europe 977

On the other hand, the recent enactment of laws promoting secularism in some European countries may be perceived by minority groups—particular religious minorities and ethnic-minority communities—as institutionalized discrimination. France’s secularity laws, for example, passed in 2004 and 2010, prohibit conspicuous religious symbols (such as head scarves and turbans) in public schools, as well as the covering of the face in public. While hats, sunglasses, and scarves are excluded from the law, burqas and niqabs are effectively banned in public places (Davis 2011). The relative absence of religious symbols in public life may make racial-ethnic and religious discrimination less overt, but minorities who are left with the choice either to constrain their participation in public life, or to modify their cultural and religious practices, may interpret these laws as discriminatory. Another important cultural difference, particularly as it relates to multiple jeopardy and multiple consciousness, concerns the increasing acknowledgment of “multiple discrimination” within public policy around the world, and particularly in the EU (Burri and Schiek 2008). Multiple discrimination refers to discrimination that is based on multiple social statuses, and includes both “additive discrimination” as described by Beal (1970) and “multiple jeopardy” as described by King (1988).7 The importance of multiple discrimination for public policy was high- lighted in the Beijing platform for Action for Equality, Development, and Peace, issued by the United Nations Fourth World Conference on Women in 1995 (United Nations 1995), and it has since been recognized explicitly within several EU policy documents and legislation (Burri and Schiek 2008; Carles and Jubany-Baucells 2010). In 2003, the European Commission launched campaigns to raise awareness of discrimination issues in all member states, and 2007 was proclaimed the Euro- pean Year of Equal Opportunities—a central focus of which was the issue of multiple discrimination (European Commission 2007:4). To the extent that they have been successful, these legislative initiatives have likely shaped perceptions of personal discrimination, as well as individuals’ understanding of discrimination more generally. Finally, while European antidiscrimination policy has made significant gains in recent decades, particularly as it concerns multiple discrimination, several studies find that racism and xenophobia have intensified over the past decade, and that perceptions of discrimination remain high, particularly among Roma populations and immigrants from outside of the EU (Amiraux and Guiraudon 2010; European Commission 2007; Zick et al. 2008).8 Data collected in 2006 from across the EU suggest that at least 50% of individuals believe that discrimination on the basis of ethnic origin, disability, and sexual orientation remains widespread in Europe. Almost half of respondents also indicated that they believed discrimination on the basis of ethnic origin and “religion or beliefs” had increased in the previous five years (European Commission 2007). It is likely that events surrounding the 9/11

7 Crenshaw (1989, 1991) and Best et al. (2014) use the term intersectional discrimination to describe this phenomenon. I follow Burri and Schiek (2008) and Sargeant (2011) in using the term multiple discrimination, as this term is increasingly used in EU policy, and is also used in the Eurobarometer. 8 Ford (2008) finds evidence that racism has declined in Britain, however. Changes in racial and ethnic prejudice are likely uneven across the 10 countries examined here. 978 Harnois attacks in New York, the expansion of the EU in 2004, the Madrid train bomb- ings in 2004, the London bombing in 2005, and the global economic recession of 2007 have all influenced patterns of discrimination, and have similarly shaped individual and cultural understandings of inequality and discrimination (Zick et al. 2008). It is within this context that the 2008 Eurobarometer was conducted, and the analyses of these data published by the European Commission (2008) suggest that a significant proportion (37%) of Europeans across the 30 countries surveyed perceive multiple discrimination as relatively widespread. The Commission’s (2008:81) report suggests that “personal experience of discrimination is a key influence on perceptions of multiple discrimination” with those who perceive having personally experienced discrimination being more likely to perceive multiple discrimination as widespread. The report also suggests that women (as opposed to men) and the young (as opposed to the elderly) are more likely to view multiple discrimination as a significant problem. The current study builds upon this research, moving from bivariate, descriptive analyses to those that are multivariate and inferential. A multivariate approach is particularly useful for incorporating intersectionality, as it allows for a simultaneous examination of the relationship among holding multiple social statuses, experiences of multiple jeopardy, and holding a consciousness of multiple discrimination.

RESEARCH QUESTIONS AND HYPOTHESES

Intersectionality argues that a consideration of multiple systems of inequality yields a more accurate understanding of how individuals perceive discrimination and how these experiences link to broader understandings of inequality. This study, therefore, asks two main questions: First, to what extent do individuals with multiple minority statuses understand their experiences with mistreatment along multiple axes of inequality? Second, to what extent does membership in multiple minority groups, along with perceptions of multiple forms of discrimination, correlate with a broader awareness of intersecting inequalities? Again, both King’s (1988) theory and intersectionality theory more generally emphasize the importance of context in shaping individuals’ experiences and consciousness of multiple inequalities. Neither hypothesize that in all social contexts, one particular social group will face the most mistreatment, or have the highest level of multiple consciousness. They do, however, oﬀer two speciﬁc hypotheses:

Hypothesis 1: Individuals with multiple minority statuses will be more likely than those with no minority statuses, or a single minority status to interpret their personal experiences with mistreatment in terms of multiple systems of inequality.

Hypothesis 2: Individuals who perceive that they themselves have experienced multiple forms of discrimination will be more likely than others to hold a multiple consciousness. Holding multiple minority statuses and experiencing multiple jeopardy will positively and signiﬁcantly correlate with awareness of multiple discrimination. To the extent that the concepts and relationships developed in King’s theory apply to a broader range of minority groups, and to groups outside of the United Intersecting Inequalities in Western Europe 979

States, this lends support to her critique of monist approaches for understanding inequality, and underscores the merits of intersectionality.

DATA AND MEASUREMENT

The Eurobarometer Survey is a multicountry, multilingual survey conducted regularly on behalf of the European Commission for the purpose of monitoring social and political attitudes in Europe. The survey includes individuals age 15 and over, and is collected using a multistage, random probability sample in each state of the EU. In each country, a number of sampling points were drawn with the probability proportional to population size (for a total coverage of the country) and to population density (European Commission 2008). Data were collected in face- to-face interviews in respondents’ homes, in the relevant national language. This study analyzes data from a subset of the 2008 Survey (296, Wave 69.1), which includes 9,012 individuals in 10 countries including the six EU founding members: France, Belgium, the Netherlands, western Germany, Italy, and Luxem- bourg, as well as Denmark, Ireland, Great Britain, and Northern Ireland. The sample size within each country varies according to the overall national population, with the smallest sample from Northern Ireland (N = 301) and the largest from France (N = 1,054). Consistent with the Survey documentation (Papacostas 2008), analyses include weights (unless otherwise noted) adjusting each national sample in proportion to its share in the total population aged 15 for this subgroup of countries.9 The main variables of interest concern respondents’ personal experiences with multiple jeopardy, measured in terms of their personal experiences with discrimination, and their multiple consciousness, measured by examining their perceptions of the pervasiveness of multiple discrimination. Unlike most large-scale surveys in the United States (Harnois and Ifatunji 2011), the Eurobarometer’s questions concerning perceived discrimination explicitly recognize that respondents may have experienced multiple types of discrimination. The survey asks, “In the past 12 months have you personally felt discriminated against or har- assed on the basis of one or more of the following grounds?...ethnic origin, gender, sexual orientation, age, religion or belief, disability?” Respondents are given a card that lists these statuses and are told explicitly that they may attribute their mistreatment to more than one social status. With this question design, those who perceive multiple types of discrimination—either simultaneously or serially—are able to frame their experiences in the context of multiple social statuses and inequalities. Respondents’ perceptions of personal experiences with discrimination are analyzed in two ways. I begin by examining the prevalence of particular forms of discrimination and the extent to which particular forms of discrimination are perceived as occurring in combination. In these analyses, I use dummy variables to indicate whether respondents perceive having experienced a particular form of discrimination (1 = yes). In later analyses, perceptions of personal discrimination are

9 Weighted analyses use STATA’s svyset command, with country speciﬁed as strata. 980 Harnois assessed with variables indicating whether respondents perceive having experienced either no form of discrimination, one form of discrimination, or multiple forms of discrimination. The extent to which respondents hold a multiple consciousness is assessed by their broader awareness of multiple discrimination. In the Eurobarometer, respondents are ﬁrst asked about their own personal experiences with discrimination, and then asked about their perceptions of discrimination more generally:

We have just been discussing discrimination based on ethnic origin, gender, sexual orientation, age, religion or belief, and/or disability. Some people may experience discrimination on the basis of more than only one of these characteristics. If they experience discrimination on several grounds, we call this “multiple discrimination.” Could you please tell me whether, in your opinion, multiple discrimination is very widespread, fairly widespread, fairly rare, or very rare in (OUR COUNTRY). Responses of “fairly rare” or “very rare” are coded 0, and “very widespread” or “fairly widespread” are coded 1.10 Additional variables include respondents’ age, gender, nation of residence, education, national background, as well as respondents’ self-identiﬁcation as a minority in terms of ethnicity, religion, disability, and sexual orientation.11 Age is measured by comparing four age groups: 15–24, 25–39, 40–54, and 55 and older. Education is assessed with a series of dummy variables indicating, for those not currently attending school, the age at which respondents’ formal schooling ended: age 14 or younger, age 15–18 (the reference group), age 19–21, and education beyond age 21. An additional dichotomous variable is included indicating whether respondents are currently attending school. Workforce status is assessed with a dichotomous variable where 1 = not currently working (and not in school), otherwise 0. National background is assessed with a question asking respondents where they “personally” were born and is grouped into three categories: the country where the interview is taking place; another country within the EU; or outside of the EU. Finally, respondents are asked about their sense of self as a member of a minority group: “Where you live, do you consider yourself to be part of any of the following? Please tell me all that apply: an ethnic minority, a religious minority, a sexual minority, a minority in terms of disability?” Responses are coded into a series of dummy variables, where 1 indicates identiﬁcation with a particular minority group, otherwise 0.

ANALYTIC STRATEGY AND RESULTS

The analysis begins with an overview of the sample, including an overview of which forms of discrimination tend to be perceived independently and in combination. Multinomial logistic regression models follow, which show how membership in particular minority groups shapes the likelihood of individuals’ perceiving a single type of discrimination, multiple types of discrimination, or no discrimination at all. The last portion of the analysis examines the relationship between multiple

10 This survey question originally had four response categories. In the countries analyzed here, 17.58% of respondents said multiple discrimination was “very rare,” 38.25% said “fairly rare,” 38.08 % said “fairly widespread,” and 6.08% said “very widespread.” 11 Analyses would ideally include a control for income, but the survey does not include income data. Intersecting Inequalities in Western Europe 981 jeopardy and consciousness. What factors are associated with perceiving multiple discrimination as widespread? And how do personal experiences of discrimination combine with social statuses to inﬂuence who is likely to perceive a high prevalence of multiple discrimination? Binary logistic regression models demonstrate that, in general, minority group membership and personal experiences with discrimination both increase the likelihood of holding a multiple consciousness.

The Prevalence of Perceived Discrimination

Table I presents an overview of the sample characteristics, with unweighted data and with missing cases omitted listwise. The means reflect the proportion of respondents who fall into particular categories as a proportion of the final sample. Women comprise 54.7% of the sample; 2.9% of respondents self-identify as ethnic minorities and less than 1% identify as a minority with respect to sexual orientation. The variables concerning discrimination are presented in the bottom portion of the table, which shows that perceptions of personal experiences with discrimination in the past year are relatively uncommon. Among the total sample, for example, less than 2% of respondents report that they have experienced discrimination based on religion or belief, and only 3.1% of respondents perceive having experienced discrimination based on gender.12 Perceptions of age discrimination are the most prevalent, with 5.2% of respondents answering that they have personally experienced this form of mistreatment. Overall, 2.5% of respondents report having experienced multiple forms of discrimination in the past year. Table II shows how perceptions of various forms of discrimination relate to one another. The numbers in the leftmost columns show the (unweighted) per- centage of individuals in the sample who attributed their mistreatment to a single social status and to multiple social statuses. The top row, for example, indicates that 36.07% of those who perceive having personally experienced ethnic discrimination also perceived another form of discrimination in the past year. More than 40% of those who perceive having experienced gender discrimination, and more than half of those who perceive having experienced discrimination based on religion or belief, also perceived having experienced discrimination based on another social status. The numbers in the right portion of Table II show the relative prevalence of particular combinations of perceived discrimination. Of the 244 individuals who perceive ethnic discrimination, for example, 27 individuals (11.7%) also perceive gender discrimination, and 51 (20.9%) perceive having experienced discrimination on the basis of religion or belief. Of those who perceive having experienced gender discrimination, nearly three in ten also perceive having experienced age-based discrimination. The table indicates that every form of discrimination examined is perceived, by at least some respondents, to have occurred in combination. The most common combinations reflect age-based and gender-based discrimination, and

12 These ﬁgures result from analyzing the entire sample. Among women respondents, 4.8% perceive gender discrimination, 25.94% of ethnic minorities perceive ethnic discrimination, and 11.9% of religious minorities perceive discrimination based on religion or belief. 982 Harnois

Table I. Descriptive Statistics (Unweighted), 2008 Eurobarometer Survey of 10 Western European Countries, N = 8,911

Mean Std. Dev.

Gender (women = 1, men = 0) 0.547 (0.50) Age 15–24 0.110 (0.31) 25–39 0.232 (0.42) 40–54 0.277 (0.45) 55+ 0.381 (0.49) Self-Described Minority Status (1 = yes) Ethnic 0.029 (0.17) Religious 0.034 (0.18) Sexual orientation 0.008 (0.09) Disability 0.022 (0.15) National Background Country of residence 0.918 (0.28) Other state outside the EU 0.031 (0.17) Other state in EU 0.052 (0.22) Currently Working 0.511 (0.50) Currently in School 0.073 (0.26) Not Working, Other 0.416 (0.49) Education Through age 14 0.130 (0.34) Through ages 15–18 0.400 (0.49) Through ages 19–21 0.180 (0.38) Beyond age 21 0.217 (0.41) Country of Residence France 0.118 (0.32) Belgium 0.112 (0.32) The Netherlands 0.114 (0.32) West Germany 0.116 (0.32) Italy 0.113 (0.32) Luxembourg 0.057 (0.23) Denmark 0.113 (0.32) Ireland 0.112 (0.32) Britain 0.112 (0.31) Northern Ireland 0.034 (0.18) Perceived Discrimination based on (1 = yes): Ethnicity 0.027 (0.16) Gender 0.031 (0.17) Sexual orientation 0.011 (0.10) Age 0.052 (0.22) Religion or belief 0.016 (0.13) Disability 0.021 (0.14) Single form of discrimination 0.099 (0.30) Multiple forms of discrimination 0.025 (0.16) Aware of multiple discrimination (1 = widespread)* 0.441 (0.50)

All variables have a minimum value of 0 and a maximum value of 1. *N for Awareness of multiple discrimination is 7,824. religion or belief with ethnic discrimination. Of those who perceive discrimination on the basis of disability, nearly 18% also perceive having experienced age-based discrimination in the past year. nescigIeulte nWsenErp 983 Europe Western in Inequalities Intersecting

Table II. Perceptions of Discrimination Based on a Single Attribution and in Combination With Other Forms of Discrimination, in 10 Western European Countries, 2008 Eurobarometer, Unweighted (N = 8,911)

Percent of Respondents Who Perceive One Form of Discrimination Who Also Perceive Discrimination Based On:

Single Multiple + Sexual + Religion or Attribution Attribution + Ethnicity + Gender Orientation + Age Belief + Disability

Ethnicity (N = 244) 63.93% 36.07% — 11.07% 7.38% 11.48% 20.90% 4.92% Gender (N = 276) 55.80% 44.20% 9.78% — 7.61% 29.71% 9.42% 8.70% Sexual Orientation 58.16% 41.84% 18.37% 21.43% — 19.39% 13.27% 7.14% (N = 98) Age (N = 515) 71.09% 28.91% 4.90% 15.70% 5.80% — 5.20% 6.20% Religion or Belief 40.97% 59.03% 35.42% 18.06% 9.03% 18.06% — 9.72% (N = 144) Disability (N = 185) 69.73% 30.27% 6.49% 12.97% 3.78% 17.84% 7.57% — 984 Harnois

Table III. Multinomial Logistic Regression of Perceived Discrimination on Selected Sociodemographic Characteristics in 10 Western European Countries. Reference Group in Dependent Variable Is No Per- ceived Discrimination in the Past Year. Eurobarometer Survey, 2008 (N = 8,911)

Model 1 Model 2

Single Multiple Single Multiple Discrimination Discrimination Discrimination Discrimination

Gender (women = 1) 0.153 0.596** 0.147 0.557* (0.111) (0.224) (0.125) (0.266) Age 25–39 0.152 0.359 0.158 0.340 (0.219) (0.369) (0.219) (0.361) Age 40–54 0.0610 0.146 0.0692 0.157 (0.145) (0.278) (0.146) (0.277) Age 55 and older –0.300 –0.164 –0.297 –0.195 (0.185) (0.335) (0.185) (0.338) Ethnic minority 1.031*** 1.347** 1.152*** 1.826*** (0.224) (0.448) (0.226) (0.438) Religious minority 0.537* 0.963* 0.654* 1.430*** (0.264) (0.400) (0.285) (0.396) Sexual orientation minority 0.962 1.121 0.966 1.368* (0.555) (0.713) (0.551) (0.590) Disabled minority 1.373*** 2.191*** 1.733*** 3.015*** (0.268) (0.404) (0.407) (0.634) Born outside the EU 0.757*** 0.687 0.747*** 0.624 (0.224) (0.450) (0.224) (0.459) Born in other EU state 0.131 –0.177 0.117 –0.218 (0.231) (0.447) (0.233) (0.442) Attending school currently –0.141 0.359 –0.143 0.348 (0.279) (0.450) (0.279) (0.448) Not working, not in school 0.360* –0.023 0.365* 0.005 (0.154) (0.252) (0.154) (0.249) Education through age 14 0.186 0.325 0.185 0.344 (0.166) (0.373) (0.167) (0.366) Education through ages 19–21 0.082 0.0270 0.084 0.0502 (0.144) (0.312) (0.145) (0.310) Education beyond age 21 0.187 0.599* 0.101 –0.04 (0.155) (0.269) (0.222) (0.444) Women with a disability –0.687 –1.533* (0.542) (0.761) Ethnic and religious minority –0.659 –2.067* (0.639) (0.839) Women with education 0.174 0.987* beyond age 21 (0.277) (0.484) Constant –2.496*** –4.352*** –2.502*** –4.380*** (0.150) (0.372) (0.151) (0.388)

Coeﬃcients reﬂect weights. Linearized standard errors in parentheses. ***p < 0.001, **p < 0.01, *p < 0.05.

Who Perceives Having Experienced Multiple Discrimination?

The next portion of the analysis investigates how social statuses, alone and in combination, structure perceptions of discrimination. Table III shows the results from two multinomial logistic regression models. In both models the dependent variable is perceptions of personal experiences of discrimination, grouped into three Intersecting Inequalities in Western Europe 985 categories: those who perceive having experienced no forms of discrimination, those who perceive having experienced a single form of discrimination, and those who perceive having experienced multiple forms of discrimination. In each model, the reference group in the dependent variable is those who perceive no personal experience with discrimination in the past year. Model 1 investigates the sociodemographic correlates of perceiving multiple sources of personal discrimination and shows that, in general, perceptions of personal discrimination are structured by gender, ethnicity, religion, (dis)ability, national background, education, and workforce status. The left column of Model 1 shows the likelihood of perceiving one form of personal discrimination, compared to the odds of perceiving no discrimination. For example, compared to those who do not identify as ethnic minorities, respondents who do identify as ethnic minorities are 2.80 times as likely to perceive a single form of discrimination as opposed to no discrimination (Exp (1.031) = 2.80). Compared to those who do not identify as religious minorities, those who describe themselves as religious minorities are 71% more likely to perceive having experienced one form of discrimination, as opposed to perceiving no personal experiences with discrimination in the past year (Exp (0.537) = 1.71). The right column of Model 1 shows the odds of perceiving two or more forms of personal discrimination, as opposed to perceiving no discrimination in the past year. Here the results show that women, ethnic and religious minorities, and people with disabilities are significantly more likely than their male, ethnic and religious majority, and nondisabled counterparts to perceive multiple sources of discrimination, as opposed to no discrimination. Compared to men, women are nearly 81% more likely to perceive multiple forms of discrimination, as opposed to no discrimination. This model suggests that, compared to people without disabilities, those who identify as disabled are more than eight times as likely to perceive having experienced multiple forms of discrimination, as opposed to experiencing no discrimination in the past year (though this effect is reduced when an interaction term is included in Model 2). Model 2 builds on the previous model by including terms that capture the intersection of multiple social statuses. Combinations of all minority statuses were tested for interactive effects and three were statistically significant. While two of the interaction terms are negative (Disability*women, Ethnic Minority*Religious minority), when interpreted in conjunction with the main effects, the overall effect of belonging to these groups is positive relative to the reference group. Table IV presents the predicted probabilities derived from Model 2, and illustrates more clearly how various combinations of social statuses shape the likelihood of individuals’ perceiving multiple forms of discrimination, one form of discrimination, or no forms of discrimination.13 As is shown in this table, among men and women born in the country in which they are residing, who are aged 40–54, currently working, and whose formal education ended between the ages of 19–21, and who report holding no minority statuses, the probability of perceiving multiple forms of discrimination is low (0.013 and 0.022, respectively). For those who hold multiple minority statuses, the odds of perceiving multiple sources of discrimination increase dramatically. Among women who identify as a

13 The predicted probabilities are calculated using the margins command in STATA (IC 12.1). 986 Harnois

Table IV. Selected Predicted Probabilities for Individuals Perceiving Personal Experiences With No Discrimination, a Single Form of Discrimination, and Multiple Forms of Discrimination in the Past 12 Months, in 10 Western European Countries, 2008 Eurobarometer, N = 8,911

No Single Multiple Discrimination Discrimination Discrimination

Men with no minority statuses 0.907 (0.012) 0.080 (0.011) 0.013 (0.004) Women with no other minority statuses 0.887 (0.012) 0.090 (0.011) 0.022 (0.005) Men with education beyond age 22 0.899 (0.017) 0.087 (0.016) 0.013 (0.004) and with no other minority statuses Women with education beyond age 22 0.833 (0.023) 0.111 (0.019) 0.055 (0.015) and with no other minority statuses Sexual minority women with no other 0.731 (0.095) 0.195 (0.084) 0.073 (0.400) minority statuses Women with a disability and no other 0.713 (0.068) 0.206 (0.060) 0.080 (0.035) minority statuses Ethnic minority women with no other 0.675 (0.055) 0.217 (0.044) 0.107 (0.043) minority statuses Religious minority women with no 0.768 (0.049) 0.150 (0.038) 0.082 (0.033) other minority statuses Religious and ethnic minority men with 0.755 (0.099) 0.209 (0.093) 0.036 (0.023) no other minority statuses Religious and ethnic minority women, 0.712 (0.066) 0.228 (0.098) 0.060 (0.005) with no other minority statuses

Unless otherwise noted, the probabilities assume respondents ages 40–54, currently working, whose formal education ended between the ages of 19–21, and who were born in the country in which they are currently residing. Probabilities shown reflect weights. Delta-method standard errors in parentheses. minority with respect to sexual orientation but who hold no other minority statuses, the probability of perceiving multiple forms of discrimination in the past year is 0.07; for religious minority women the probability is approximately 0.08. For ethnic minority women the probability of perceiving multiple forms of discrimination in the past year increases to 0.107. The probability of having perceived multiple forms of discrimination is 0.06 among women who identify as both religious and ethnic minorities and 0.08 among women with disabilities (assuming no other minority statuses). While Model 1 found a significant effect for high levels of educational attainment on perceptions of multiple discrimination (b = 0.599), Model 2 shows that this effect is moderated by gender. Compared to women with no other minority statuses whose education ended between the ages of 19–21, women whose schooling ended beyond age 21 were significantly more likely to perceive multiple discrimination in the past year (0.055, as compared to 0.022). Ceteris paribus, men whose education ended beyond age 21 were no more likely to perceive multiple discrimination than were men with lower levels of education.

Who Perceives Multiple Discrimination as Widespread?

The ﬁnal portion of the analysis investigates the link between individuals’ holding multiple minority statuses, perceiving personal experiences with multiple forms of discrimination, and awareness of multiple discrimination as a widespread social Intersecting Inequalities in Western Europe 987

Table V. Binary Logistic Regression of Awareness of Multiple Discrimination on Selected Sociodemo- graphic Characteristics in 10 Western European Countries, 2008 Eurobarometer (N = 7,824)

Model 1 Model 2 Model 3

Gender (women = 1) 0.350*** 0.330*** 0.326*** (0.071) (0.072) (0.071) Age 25–39 0.168 0.175 0.161 (0.145) (0.146) (0.145) Age 40–54 –0.0504 –0.0617 –0.0619 (0.096) (0.097) (0.096) Age 55 and older –0.516*** –0.503*** –0.504*** (0.112) (0.113) (0.112) Ethnic minority 0.328 0.204 0.181 (0.201) (0.203) (0.200) Religious minority 0.657*** 0.555** 0.572** (0.196) (0.197) (0.192) Sexual orientation minority 1.972*** 1.888*** 1.842*** (0.573) (0.560) (0.553) Disabled minority 0.487* 0.127 0.245 (0.232) (0.238) (0.232) Born in other EU state –0.318+ –0.348+ –0.340+ (0.182) (0.188) (0.188) Born outside the EU 0.250 0.145 0.161 (0.200) (0.206) (0.204) Attending school currently –0.133 –0.132 –0.140 (0.181) (0.182) (0.182) Not working, not in school 0.121 0.0949 0.106 (0.0951) (0.0955) (0.0954) Education through age 14 0.161 0.157 0.150 (0.112) (0.113) (0.113) Education through ages 19–21 0.0171 0.0235 0.0175 (0.0995) (0.101) (0.100) Education beyond age 21 –0.0359 –0.0523 –0.0580 (0.096) (0.097) (0.097) Ethnic discrimination 0.506* (0.222) Gender discrimination 0.516* (0.208) Sexual orientation discrimination –0.402 (0.301) Age discrimination 0.469** (0.162) Religious discrimination 0.921** (0.283) Disability discrimination 1.119*** (0.249) Single form of personal discrimination 0.635*** (0.110) Multiple forms of personal discrimination 1.077*** (0.252) Constant –0.235* –0.285** –0.292** (0.0961) (0.0964) (0.0963)

Standard errors in parentheses; coeﬃcients and standard errors reﬂect weights. ***p < 0.001, **p < 0.01, *p < 0.05. phenomenon. Recall that survey respondents are told the precise meaning of multiple discrimination in the context of the survey and are then asked whether they believe this is relatively rare or widespread. 988 Harnois

Table V shows the results from three binary logistic regression models, where the dependent variable is perceptions of multiple discrimination as relatively rare (coded 0) or widespread (coded 1). Model 1 includes respondents’ sociodemographic characteristics, as well as respondents’ sense of her- or himself as a member of a minority group.14 Compared to those who consider themselves to be part of the majority, those who identify as a minority with respect to religion, sexual orientation, or disability are more likely to see multiple discrimination as widespread. Compared to men, women are more than 40% more likely to perceive multiple discrimination as widespread, controlling for other factors (Exp (0.35) = 1.42). Those who identify as religious minorities are nearly twice (1.92 times) as likely as their nonreligious minority counterparts to view multiple discrimination as widespread. Not all minority statuses increase awareness of multiple discrimination, however. Ethnic minority status is statistically nonsignificant, and compared to the youngest group in the sample, respondents aged 55 and above are significantly less likely to view multiple discrimination as widespread.15 Models 2 and 3 build on the previous model by including perceptions of personal discrimination. Results show that after controlling for social statuses, personal experiences with discrimination based on gender, ethnicity, disability, age, and religion each increase the likelihood of respondents’ perceiving multiple discrimination as widespread. With the exception of disability status, all of the minority statuses that were significant in Model 1 remain significant when perceptions of discrimination are included, which demonstrates how social statuses work with experiences of discrimination to shape multiple consciousness. Other things being equal, women are significantly more likely than men to perceive multiple discrimination as widespread, for example, but women who have had personal experiences with gender discrimination are even more likely than other women (and also more likely than men who perceive gender discrimination) to perceive multiple discrimination as widespread. Model 3 shows a more general model of the relationship among social statuses, personal experiences with single and multiple discrimination, and awareness of multiple discrimination. Compared to those who perceive experiencing no discrimination, those who perceive having experienced a single form of discrimination are nearly twice as likely (88% more likely) to view multiple discrimination as widespread. Personal experiences with “single jeopardy” thus increase the likelihood of holding a multiple consciousness. Personal experiences with multiple jeopardy have an even greater impact. Again, compared to those who perceive experiencing no discrimination, those who report that they themselves have experienced multiple forms of discrimination are nearly three times as likely (193% more likely) to view multiple discrimination as a widespread phenomenon.

DISCUSSION

As a theoretical perspective, intersectionality draws attention to the dynamic and contextual relationships among multiple systems of inequality as they exist

14 I also tested for interacting social statuses, but none were significant, so are not included here. 15 It is unclear from this analysis whether this finding reflects age or cohort differences. Intersecting Inequalities in Western Europe 989 within, and across, multiple levels of society (Berger and Guidroz 2009; Choo and Ferree 2010; McCall 2005; Winker and Degele 2011; Yuval-Davis 2006). Scholars of intersectionality argue that individuals with multiple minority statuses often face mistreatment that stems from multiple, intersecting systems of inequality (Crenshaw 1989, 1991; Essed 1991; Hill Collins 2000; Weber 2010). King (1988) argues more specifically that experiences with “multiple jeopardy” facilitate the development of “multiple consciousness.” Although based on the history, politics, and experiences of African American women, scholars have used these concepts to understand a range of inequalities as experienced by marginalized groups around the world. This study is the first of its kind to link intersectional theories of multiple jeopardy and multiple consciousness with large-scale survey data. While a single study is rarely sufficient for proving or disproving a theory, the results of this study provide support for intersectionality theory in two different respects. First, of those respondents who perceived discrimination, many indicated that they had personally experienced multiple forms of discrimination in the past year (Table II). Nearly a third of those who perceived experiencing ethnic discrimination also perceived experiencing discrimination based on another social status. More than 40% of those who perceived experiencing gender discrimination also reported experiencing another form of discrimination, and more than half of those who perceived experiencing discrimination based on religion or beliefs, also perceived discrimination based on a second social status. When asked about their experiences with discrimination, and when given the opportunity to attribute their mistreatment to multiple social statuses, a substantial proportion of respondents frame their experiences in terms of multiple inequalities. More specifically, the results show that the likelihood of perceiving multiple forms of discrimination, as opposed to no discrimination, is higher among individuals with multiple minority statuses (Tables III and IV). Women and men who hold no other minority statuses are among the least likely to perceive experiencing multiple discrimination. Religious minority women, ethnic minority women, and women with disabilities are all among the most likely to perceive having experienced multiple discrimination as opposed to no discrimination in the past year. These findings align well with Hypothesis 1, which predicted that individuals with multiple minority statuses would be more likely than others to interpret their personal experiences of mistreatment in terms of multiple systems of inequality. Other things being equal, women with high levels of education are more likely than women with low to mod- erate levels of education to perceive multiple discrimination—a finding not anticipated by intersectionality theory. This finding is, however, consistent with some existing research on perceived discrimination (Gee et al. 2007; Kessler et al. 1999; Sigelman and Welch 1991). Hypothesis 2 concerned the relationship between perceptions of multiple jeopardy and consciousness, and here, too, the results are generally consistent with intersectionality. King argued that those who hold multiple minority statuses experience mistreatment stemming from multiple social inequalities, and are more aware of intersecting systems of inequality, compared to those who hold one or no minority statuses. The analyses presented here show that social statuses and personal experiences with discrimination both play a role in predicting multiple consciousness. Individuals who hold multiple minority statuses and who perceive multiple 990 Harnois forms of discrimination are more likely than others to perceive multiple discrimination as relatively widespread (Table V). Consistent with the European Commis- sion’s descriptive analyses of the Eurobarometer data, but not explicitly anticipated within the framework of intersectionality, is the finding that personal experiences with mistreatment based on a single dimension of inequality (“single jeopardy”) also significantly predict multiple consciousness (though not to the same extent as experiences with multiple jeopardy). In addition to providing support for King’s theory, findings underscore the importance of social context for theorizing and analyzing intersecting inequalities. In the United States, intersectional scholarship has focused primarily on race, gender, class, and sexuality (Choo and Ferree 2011; Knapp 2005; Weber 2010). The present study affirms the importance of these inequalities in Western Europe, but adds to this the important—and often neglected—statuses of religion and disability. In the subsample of the Eurobarometer analyzed here, approximately 30% of those who perceived discrimination based on disability also perceived experiencing some other form of discrimination within the past year, as did nearly 60% of those who perceived discrimination based on religion or belief (Table II). Intersecting oppressions are “organized through diverse local realities” (Hill Collins 2000:228), and these findings highlight the importance of contextual flexibility in the deployment of intersectionality theory and analysis (Choo 2012; Ferree 2009; Knapp 2005).

Limitations

While the Eurobarometer’s unique questions concerning discrimination enable an analysis of multiple jeopardy and consciousness, they also introduce some nota- ble limitations. When asked about their personal experiences with discrimination, respondents are told explicitly that they may attribute their mistreatment to more than one social status, and 2.7% of respondents select multiple attributions (Table I). Given the wording of the survey question, however, it remains unclear whether these respondents perceive having experienced multiple forms of discrimination serially (e.g., gender discrimination in one context, and ethnic discrimination in another context) or simultaneously (e.g., a single act of gendered racism). The concept of multiple jeopardy includes both, but clearly emphasizes the latter. Thus, there is a potential discrepancy between how multiple jeopardy is conceptualized by King and how it is assessed here. Another limitation is that the survey prompts respondents with the definition of multiple discrimination prior to asking them about its prevalence. While this question wording helps to ensure that respondents understand the question before answering, it also likely introduces some bias, increasing the per- centage of people who perceive multiple discrimination as widespread. A final limitation is that while intersectionality argues that lived experiences with multiple jeopardy lead to the development of multiple consciousness, the analyses here identify significant correlations, but not necessarily causation. It may be that individuals who already hold a multiple consciousness are more likely to perceive their own experiences of mistreatment in terms of multiple inequalities. These Intersecting Inequalities in Western Europe 991 limitations raise important questions for future research and, as discussed below, underscore the need for large-scale surveys to address intersecting inequalities and multiple discrimination in particular.

CONCLUSION

Despite these limitations, the findings from this study are consistent with King’s theory of multiple jeopardy and consciousness, and with intersectionality more generally. Many people who perceive mistreatment frame their experiences in terms of multiple inequalities, and those who do are more likely than others to believe that multiple discrimination is widespread. Although the Eurobarometer’s questions concerning personal experiences with discrimination and awareness of multiple discrimination may not map seamlessly onto the concepts of multiple jeopardy and multiple consciousness, the analyses here nonetheless provide strong support for the relationships that intersectionality theory predicts. Beyond this, the findings demonstrate the importance of centralizing intersectionality, and multiple discrimination more specifically, in future social science research as well as in public policy. As mentioned previously, the 1997 Amsterdam Treaty, along with the Racial Equality and Employment Framework Directives of 2000, and the declaration of 2007 as the European Year of Equal Opportunities, all brought attention to issues of inequality and discrimination, and to multiple discrimination in particular. It is the European Commission’s acknowledgment of multiple discrimination that is largely responsible for the inclusion of a question focusing explicitly on multiple discrimination in the 2008 Eurobarometer (Burri and Schiek 2008). The Euro- barometer is, after all, a survey conducted on behalf of the European Commission. While intersectionality theory originated in the United States, multiple discrimination is not currently recognized as such within the U.S. legal frameworks, nor is multiple discrimination acknowledged in most U.S. social science survey research (Crenshaw 1989; Harnois 2014). Few large-scale U.S. surveys ask respondents about such a wide range of minority identities, such varied forms of discrimination (e.g., religious, sexual orientation, disability), and encourage them to select “all that apply.” To my knowledge, there are no existing large-scale surveys of the United States that explicitly address multiple discrimination. More surveys of this type are needed, in the United States, in Europe, and elsewhere. For, without such surveys, it is nearly impossible to document the prevalence and correlates, and consequences of multiple discrimination. Writing of the importance of intersectionality for conceptualizing discrimination against African American women, Crenshaw (1991:1244) notes: “[M]any of the experiences Black women face are not subsumed within the traditional boundaries of race or gender discrimination as these boundaries are currently understood.... [T]he intersection of racism and sexism factors into Black women’s lives in ways that cannot be captured wholly by looking at the race or gender dimensions of those experiences separately.” Focusing on Western Europe, the analyses here cannot speak directly to multiple jeopardy as experienced by African American women. 992 Harnois

The analyses do lend strong statistical evidence to the importance of conceptualizing and analyzing for discrimination within an intersectional framework. Future research is needed to more fully assess multiple jeopardy, consciousness, and discrimination in diverse social contexts. The results here suggest that this research may well yield important insights into how marginalized groups perceive discrimination, and the consequences of this mistreatment for shaping individuals’ consciousness of inequality.

REFERENCES

Albert, M., H. Sklar, L. Cagan, N. Chomsky, R. Hahnel, M. King, and L. Sargent. 1986. Liberating Theory. Cambridge, MA: South End Press. Amiraux, Valerie and Virginie Guiraudon. 2010. “Discrimination in Comparative Perspective: Policies and Practices.” American Behavioral Scientist 53: 12: 1691–1714. Baca Zinn, Maine and Bonnie Thornton Dill. 1996. “Theorizing Diﬀerence from Multiracial Feminism.” Feminist Studies 22: 2: 321–333. Beal, Francis. 1970. “Double Jeopardy: To Be Black and Female.” In R. Morgan (ed.), Sisterhood Is Powerful: pp. 340–352. New York: Random House. Berger, Michele Tracy. 2004. Workable Sisterhood. Princeton, NJ: Princeton University Press. Berger, Michele Tracy and Kathleen Guidroz. 2009. The Intersectional Approach: Transforming the Acad- emy Through Race, Class, and Gender. Chapel Hill: University of North Carolina Press. Best, Rachel Kahn, Lauren B. Edelman, Linda Hamilton Krieger, and Scott R. Eliason. 2011. “Multiple Disadvantages: An Empirical Test of Intersectionality Theory in EEO Litigation.” Law & Society Review 45: 4: 991–1025. doi:10.1111/j.1540-5893.2011.00463.x. Bleich, Erik. 2011. The Freedom to Be Racist? How the United States and Europe Struggle to Preserve Freedom and Combat Racism. New York: Oxford University Press. Bowleg, L. 2008. “When Black + Lesbian + Woman 6¼ Black Lesbian Woman: The Methodological Chal- lenges of Qualitative and Quantitative Research.” Sex Roles 59: : 312–325. Burri, Susanne and Dagmar Schiek. 2008. Multiple Discrimination in EU Law: Opportunities for Legal Responses to Intersectional Gender Discrimination?. Brussels, Belgium: European Commission. Carles, Isabelle and Olga Jubany-Baucells (eds.). 2010. “Genderace: The Use of Racial Antidiscrimina- tion Laws: Gender and Citizenship in a Multicultural Context.” Retrieved January 24, 2013 (http://genderace.ulb.ac.be/) Central Intelligence Agency (CIA). 2009. The World Factbook 2009. Washington, DC: Central Intelli- gence Agency. Retrieved August 17, 2015 (https://www.cia.gov/library/publications/the-world-factbook/index.html) Choo, Hae Yeon. 2012. “The Transnational Journey of Intersectionality.” Gender & Society 26: 1: 40–45. Choo, Hae Yeon and Myra Marx Ferree. 2010. “Practicing Intersectionality in Sociological Research: A Critical Analysis of Inclusions, Interactions and Institutions in the Study of Inequalities.” Sociological Theory 28: 2: 129–149. Cooke, Miriam. 2001. Women Claim Islam: Creating Islamic Feminism Through Literature. New York: Routledge. Crenshaw, Kimberle. 1989. “Demarginalizing the Intersection of Race and Sex: A Black Feminist Critique of Antidiscrimination Doctrine, Feminist Theory, and Anti-Racial Politics.” University of Chicago Legal Forum 140: : 139–167. Crenshaw, Kimberle. 1991. “Mapping the Margins: Intersectionality, Identity Politics, and Violence Against Women of Color.” Stanford Law Review 43: 6: 1241–1299. Davis, Britton D. 2011. “Lifting the Veil: France’s New Crusade.” Boston College International and Com- parative Law Review 34: 1. Retrieved August 12, 2015 (http://lawdigitalcommons.bc.edu/iclr/vol34/ iss1/6) Essed, Philomena. 1991. Understanding Everyday Racism: An Interdisciplinary Theory. Newbury Park, CA: Sage. European Commission. 2007. “Discrimination in the European Union.” Retrieved December 22, 2014 (http://ec.europa.eu/justice/discrimination/ﬁles/eurobarometer_263_sum_en.pdf) European Commission. 2008. “Discrimination in the European Union: Perceptions, Experiences and Attitudes.” Retrieved December 27, 2014 (http://ec.europa.eu/public_opinion/archives/ebs/ebs_296_ en.pdf) Intersecting Inequalities in Western Europe 993

Federal Statistical Office of Germany. 2011. “Census: Just Under a Quarter of Foreigners Originate From Turkey.” Retrieved April 10, 2015 (www.destatis.de/EN/PressServices/Press/ pr/2014/04/ PE14_135_121.html) Ferree, Myra Marx. 2009. “Inequality, Intersectionality, and the Politics of Discourse.” In Emanuela Lombardo, Meier, Petra, and Verloo, Mieke (eds.), The Discursive Politics of Gender Equality: Stretch- ing, Bending and Policymaking: pp. 86–104. New York: Routledge. Ford, Robert. 2008. “Is Racial Prejudice Declining in Britain?” British Journal of Sociology 59: 4: 609– 636. Garstka, T., M. Hummert, and N. Branscombe. 2005. “Perceiving Age Discrimination in Response to Intergenerational Inequity.” Journal of Social Issues 61: 2: 321–342. Gee, Gilbert C., Aliza K. Pavalko, and J. Scott Long. 2007. “Age, Cohort, and Perceived Discrimination: Using the Life Course to Assess Self-Reported Age Discrimination.” Social Forces 86: 1: 265–290. Guenther, Katja M., Sadie Pendaz, and Fortunata Songora Makene. 2011. “The Impact of Intersecting Dimensions of Inequality and Identity on the Racial Status of Eastern African Immigrants.” Sociolog- ical Forum 26: 1: 98–120. Hackett, Conrad. 2015. “5 Facts About the Muslim Population in Europe.” Retrieved April 10, 2015 (http://www.pewresearch.org /fact-tank/2015/01/15/5-facts-about-the-muslim-population-in-europe/) Harnois, Catherine E. 2014. “Are Perceptions of Discrimination Unidimensional, Oppositional, or Inter- sectional? Examining the Relationship Among Perceived Racial-Ethnic, Gender, and Age Discrimina- tion.” Sociological Perspectives 57: 4: 470–487. Harnois, Catherine E. and Mosi A. Ifatunji. 2011. “Gendered Measures, Gendered Models: Toward an Intersectional Analysis of Interpersonal Racial Discrimination.” Ethnic and Racial Studies 34: 6: 1006–1028. Hill Collins, Patricia. 2000. Black Feminist Thought. New York: Routledge. Kessler, R.C., K.D. Mickelson, and D.R. Williams. 1999. “The Prevalence, Distribution, and Mental Health Correlates of Perceived Discrimination in the United States.” Journal of Health and Social Behavior 40: 3: 208–230. King, Deborah K. 1988. “Multiple Jeopardy, Multiple Consciousness: The Context of a Black Feminist Ideology.” Signs 14: 2: 42–72. Knapp, Gudrun-Axeli. 2005. “Race, Class, Gender: Reclaiming Baggage in Fast Travelling Theories.” European Journal of Women’s Studies 12: 3: 249–265. Kobrynowicz, Diane and Nyla R. Branscombe. 1997. “Who Considers Themselves Victims of Discrimi- nation? Individual Difference Predictors of Perceived Gender Discrimination in Women and Men.” Psychology of Women Quarterly 21: 3: 347–363. Madibbo, Amal Ibrahim. 2007. “Race, Gender, Language and Power Relations: Blacks Within Franco- phone Communities in Ontario, Canada.” Race, Class and Gender 14: 1–2: 213–226. McCall, Leslie. 2005. “The Complexity of Intersectionality.” Signs 30: 3: 1771–1800. Papacostas, Antonis. 2008. Eurobarometer 69.1: Discrimination, Radioactive Waste, and Purchasing in the European Union, February–March 2008. ICPSR25163-v2. Cologne, Germany: GESIS/Ann Arbor, MI: Inter-university Consortium for Political and Social Research [distributors]. Retrieved March 22, 2011 (http://doi.org/10.3886/ICPSR25163.v2) Risman, Barbara J. and Pallavi Banerjee. 2013. “Kids Talking About Race: Tween-agers in a Post-Civil Rights Era.” Sociological Forum 28: 2: 213–235. Roscigno, Vincent J. 2007. The Face of Discrimination. New York: Rowman & Littlefield. Roscigno, Vincent J., Sherry Mong, Reginald Byron, and Griff Tester. 2007. “Age Discrimination, Social Closure, and Employment.” Social Forces 86: 1: 314–334. Sargeant, Malcolm. 2011. Age Discrimination and Diversity: Multiple Discrimination From an Age Per- spective. Cambridge, UK: Cambridge University Press. Sigelman, Lee and Susan Welch. 1991. Black Americans’ Views of Racial Inequality: The Dream Deferred. New York: Cambridge University Press. Smith, Barbara, Beverly Smith, and Demita Frazier. 1981. “A Black Feminist Statement.” In C. Moraga and Anzaldua, G. (eds.), This Bridge Called My Back: pp. 210–218. Watertown, MA: Persephone Press. Swim, Janet K., Lauri L. Hyers, Lauri L. Cohen, and Melissa J. Ferguson. 2001. “Everyday Sexism: Evi- dence for Its Incidence, Nature, and Psychological Impact from Three Daily Diary Studies.” Journal of Social Issues 57: 1: 31–53. doi:10.1111/0022-4537.00200. United Nations. 1995. “Beijing Declaration and Platform for Action.” Retrieved August 17, 2015 (http:// www.un.org/womenwatch/daw/beijing/beijingdeclaration.html) U.S. Census Bureau. 2011. “Overview of Race and Hispanic Origin: 2010.” Retrieved August 17, 2015 (http://www.census.gov/prod/cen2010/briefs/c2010br-02.pdf) 994 Harnois

U.S. Equal Employment Opportunity Commission (EEOC). 2010. “Discrimination by Type.” Retrieved December23, 2014 (http://www.eeoc.gov/laws/types/index.cfm) U.S. Housing and Urban Development (HUD). 2015. “Fair Housing Laws and Presidential Executive Orders.” Retrieved April 24, 2015 (http://portal.hud.gov/hudportal/HUD?src=/program_of- ﬁces/fair_housing_ equal_opp/FHLaws) Ward, Jane. 2004. “‘Not All Diﬀerences Are Created Equal’: Multiple Jeopardy in a Gendered Organiza- tion.” Gender & Society 18: 1: 82–102. doi:10.1177/0891243203259503. Weber, Lynn. 2010. Understanding Race, Class, Gender, and Sexuality. New York: Oxford University Press. Welch, Susan, Lee Sigelman, Timothy Bledsoe, and Michael Combs. 2001. Race and Place: Race Rela- tions in an American City. Cambridge, UK: Cambridge University Press. Wilcox, Clyde. 1997. “Racial and Gender Consciousness Among African-American Women: Sources and Consequences.” Women & Politics 17: 1: 73–94. Winker, Gabriele and Nina Degele. 2011. “Intersectionality as Multi-Level Analysis: Dealing With Social Inequality.” European Journal of Women’s Studies 18: 1: 51–66. Yuval-Davis, Nira. 2006. “Intersectionality and Feminist Politics.” European Journal of Women’s Studies 13: 3: 193–209. Zick, Andreas, Thomas F. Pettigrew, and Ulrich Wagner. 2008. “Ethnic Prejudice and Discrimination in Europe.” Journal of Social Issues 64: 2: 233–251. doi:10.1111/j.1540-4560.2008.00559.x.